CRISPR Interference Limits Horizontal Gene Transfer in Staphylococci by Targeting DNA

Transcription

1 Supporting Online Material for CRISPR Interference Limits Horizontal Gene Transfer in Staphylococci by Targeting DNA Luciano A. Marraffini and Erik J. Sontheimer* *To whom correspondence should be addressed. This PDF file includes: Published 19 December 2008, Science 322, 1843 (2008) DOI: /science Materials and Methods Figs. S1 to S4 Table S1 References

2 Supporting Online Material Materials and Methods Bacterial strains and growth conditions S. epidermidis [RP62a (S1), LAM104 and ATCC (S2)] and S. aureus RN4220 (S3) strains were grown in BHI and TSB media, respectively. When required, the medium was supplemented with antibiotics as follows: neomycin (15 µg/ml) for selection of S. epidermidis RP62a and LAM104; tetracycline (5 µg/ml) for selection of S. epidermidis ATCC 12228; chloramphenicol (10 µg/ml) for selection of plm6- and pc194-based plasmids; and mupirocin (5 µg/ml) for selection of pg0400-based plasmids. Isopropyl β-d-thiogalactopyranoside (IPTG) was added at 1 mm when indicated. E. coli DH5α cells were grown in LB medium, supplemented with ampicillin (100 µg/ml) when necessary. Plasmid and strain construction S. epidermidis chromosomal DNA was extracted with the Wizard Genomic DNA Purification Kit (Promega). PCR reactions were performed using Pfu polymerase (Stratagene). Primers used in this study are listed in Table S1. To attain plasmid-based, IPTG-inducible expression in staphylococci, we constructed the vector plm6. The BamHI fragment of pmutinha (S4) containing the spac promoter and laci gene was inserted into the unique BamHI site present in pos1 (S5). P1/P2 (CRISPR 5 flanking sequence) and P3/P4 (CRISPR 3 flanking sequence) PCR products were cut with KpnI, ligated and inserted into pkor1 (S6), generating plm301 for deletion of S. epidermidis RP62a CRISPR sequences. P15/P16 (nes target 5 flanking sequence) and P17/P18 (nes target 3 flanking sequence) PCR products were joined by nested PCR and inserted into pkor1 to obtain plm302; this plasmid was used for mutation of the pg0400 nes sequence to generate pg0(mut), which contains nine silent point mutations in the spc1 target sequence. [Although previous studies have established that smaller numbers of spacer/target mismatches can prevent interference (S7-S10), we engineered the maximum number of silent mutations to ensure full disruption of CRISPR recognition.] Insertion of the P37/P6 (CRISPR sequences) PCR product into plm6 via KpnI sites generated pcrispr. Insertion of P19/P6 (CRISPR and leader sequences) PCR product into plm6 via KpnI sites generated pcrispr-l. To obtain pg0(i2), containing an insertion of Twort phage orf142-i2 intron (S11) in the pg0400 nes target sequence, we constructed plm313. P15/P51 (nes target 5 flanking sequence) and P52/P53 (Twort phage orf142-i2 intron), as well as P52/P53 and P54/P18 (nes target 3 flanking sequence) PCR products, were joined by nested PCR to generate the fragments P15/P53 and P52/P18. These DNA fragments were cut with BlpI, ligated and inserted into pkor1. P52 introduced GC to AT mutations in positions of the intron sequence to maintain the structure of the P1 helix (S11), which involves base-pairing between intron and 5 exon bases. For the generation of pg0(di2), containing an inactive intron in the nes target sequence of pg0400, plm319 was constructed by site-directed mutagenesis using primers P76/P77 and plm313 as template. Insertion of the P70/P71 (pg0400 nes target and 200 bp flanking sequences) PCR product into pc194 (S12) via HindIII sites generated pnes(wt-d) and pnes(wti). Insertion of P70/P71 [pg0(mut) nes target and 200 bp flanking sequences] PCR product into pc194 via HindIII sites generated pnes(mut-d) and pnes(mut-i). plm6, plm301, plm302, pcrispr, pcrispr-l, plm313 and plm319 were cloned in E. coli strain DH5α. pnes(wt) and pnes(mut) were cloned in S. aureus strain OS2 (S5). The DNA sequences of all inserts were 1

3 verified by sequencing. Strains were constructed by allelic exchange using pkor1-based plasmids, following the published protocol (S6). RT-PCR, primer extension and sequencing For RNA extraction, S. epidermidis cells were grown in 10 ml of BHI to an OD 600 of 1. Total RNA was isolated using the FastRNA Pro Blue Kit (Qbiogene) and FastPrep instrument (Qbiogene) for 35 seconds at a speed setting of 6.5. Primers P67, P68 and P69 were labeled using γ 32 P-ATP and T4 polynucleotide kinase (New England Biolabs) and PAGE-purified. Primer extension and RT-PCR reactions were performed as described in (S13) using M-MLV reverse transcriptase (Amersham). For detection of pcrispr transcripts, reverse transcription was performed using P6 and PCR using P6/P37. For detection of pg0400 transcripts, reverse transcription was performed using P8 and PCR using P8/P72. P68 primer extension product was mapped by comparison with a DNA sequencing ladder generated by the Sanger method using Klenow fragment (New England Biolabs), P68 oligonucleotide primer, and pcrispr-l DNA template as described (S14). Conjugation Conjugation was carried out by filter mating as described elsewhere (S15). We chose pg0400 for this study because it provides a suitable marker (mupirocin resistance) for selection in S. epidermidis RP62a and ATCC recipients. Overnight donor and recipient cultures growing in BHI medium at 37 C with the necessary antibiotics were diluted 1/100 and incubated until they reached an OD 600 value of approximately µl of donor and recipient were then diluted in 5ml of fresh BHI and vacuum-filtered through 0.45 µm filters (Millipore). Filters were incubated on BHI agar plates at 37 C for 18 hours and bacteria resuspended in 3 ml of BHI. Serial dilutions were then plated in BHI agar containing the appropriate antibiotic for the selection of recipients or transconjugants. The conjugation efficiency was calculated as the ratio of transconjugants to recipients. The mean of three independent experiments is reported. The genetic identity of transconjugants was corroborated by PCR analysis using primer pairs P13/P14 to amplify a gene present exclusively in S. epidermidis ATCC [SE0084, (S2)], P27/P28 to amplify a gene present exclusively in S. epidermidis RP62a [SERP1525 (S1)], P70/P71 to amplify a pg0400-specific DNA sequence and P156/P157 to amplify plm6 inserts. To distinguish between wild-type and mutant pg0400, the P70/P71 PCR product was digested with SphI, a restriction site present only in the pg0(mut) amplicon. Transformation For transformation of S. epidermidis strains, DNA was prepared from exponentially growing S. aureus cells (to minimize chromosomal DNA contamination) using a QIAGEN Plasmid Midi Kit. To eliminate chromosomal DNA, plasmid DNA was gel-purified using the QIAquick Gel Extraction Kit. DNA concentration was determined by measuring absorbance at 260 nm. Transformation was carried out as described in (S16), except that cells were recovered in 300 µl of BHI overnight. Transformation efficiency was calculated as the number of colony forming units (cfu) per µg of DNA. The genetic identities of the transformants were corroborated by PCR as described above for conjugation assays. The presence of pnes(wt) and pnes(mut) in the transformants was corroborated by P86/P87 PCR followed by digestion with SphI, which cuts the pnes(mut) amplicon but not that of pnes(wt). 2

4 Supplementary Figure S1. (A) Sequences of repeats and spacers in the RP62a CRISPR locus. Four 36-nucleotide direct repeats (DR1-4) flank three 34- or 35-nucleotide spacer sequences (spc1-3). The spc1 sequence is homologous to a region present in all staphylococcal conjugative plasmids sequenced so far. The spc2 sequence is found in the staphylococcal phage PH15, whereas the spc3 sequence has no matches in Genbank. Asterisks indicate nucleotides that are conserved among all four repeats. (B) Conservation of nes target sequences. The nucleotide 3

5 sequence of the nes gene of conjugative plasmids pusa03 (S17), pg01 (S18), psk41 (S19), plw043 (S20) and pg0400 (S15) is more than 99% identical, with no mismatches in the vicinity of the spc1 target sequence (highlighted in yellow). In contrast, pv030-8 shares only 91% identity with the rest of the sequenced staphylococcal conjugative plasmids. It carries six substitutions in the vicinity of spc1 target (indicated in red): two in the target sequence itself, one upstream and three downstream. Sequences corresponding to the accession numbers EA and AR found in the NCBI Patent Sequence Database that contain spc1 target sequence are also included in the alignment, although their precise nature and origin are not well characterized. (C) Conjugation assays using S. epidermidis ATCC 12228, a strain that lacks a CRISPR locus, as recipient. Conjugation was carried out by filter mating in triplicate; the cfu/ml values (mean +/- SD) obtained for recipients and transconjugants are shown. Recipient strains, their complementing plasmids and the donor conjugative plasmids are indicated. Conjugation efficiency (Conj. Eff.) was calculated as the recipients/transconjugants ratio. (D) pcrisprbased CRISPR expression. RT-PCR was performed using primers that span the repeat-spacer region of the S. epidermidis RP62a CRISPR locus. A PCR product of the correct size was observed in LAM104 cells carrying pcrispr only in the presence of IPTG, but not in cells harboring the empty plasmid plm6. A less intense product was obtained using RP62a total RNA. 4

6 Supplementary Figure S2. Detection of an spc1 crrna by primer extension. (A) Arrows indicate the priming sites of the sense (S, P68 in Supplementary Table S1) and antisense (A, P69) primers within spc1 of the CRISPR locus. DR denotes a CRISPR direct repeat. (B) The antisense primer shown in (A) yields a ~40-nt product (denoted by the filled arrowhead on the right) specifically in CRISPR-positive strains. All reactions contained a primer ( r ) complementary to 5S rrna as an internal positive control (P67), and the 26-nt 5S rrna extension product is indicated by the open arrowhead on the right. Lanes marked r contained the 5S rrna primer only, and lanes marked A and S also contained the antisense or sense primers, respectively. A dideoxy sequencing ladder is included to the left, and the sequences of spc1 (yellow box) and DR1 (unfilled box) are indicated. (C) CRISPR precursor RNA is cleaved at the base of a potential stem-loop structure in each repeat, consistent with a previous report (S8). 5

7 Supplementary Figure S3. Splicing of an intron-containing nickase gene is required for nickase activity. (A) Disruption of the nes sequence with a self-splicing intron. A group I selfsplicing intron (orf142-i2) from the staphylococcal phage Twort, or a three-nucleotide deletion mutant that removes the intron s essential guanosine binding site, were introduced into the middle of the nes target sequence (highlighted), generating the conjugative plasmids pg0(i2) and pg0(di2), respectively. (B) Transfer of the plasmids described above was tested using wild-type S. epidermidis RP62a as recipient. High conjugation efficiencies were obtained for pg0(i2) but not for pg0(di2). (C) Splicing of nes mrna and pre-mrna. RT-PCR was performed using total RNA from the indicated strains as template for amplification with a primer pair that spans the site of intron insertion in pg0(i2) and pg0(di2). Amplification of spliced and unspliced RNAs yields products of 587 and 839 bp, respectively. PCR reactions with plasmid DNA templates were included as amplification controls. 6

8 Supplementary Figure S4. Mixed Transformations. (A) The plasmids pnes(wt-d), pnes(wti), pnes(mut-d), and pnes(mut-i) containing wild-type and mutant nes target sites (see Figure 3A) were used to transform S. epidermidis in pair-wise combinations to allow for internally controlled transformation experiments. PCR primers used to amplify the target region in the transformants are shown at the bottom. The mutations in the pnes(mut) plasmids (highlighted in grey) include one that introduces a diagnostic SphI restriction site (underlined) that is absent from the pnes(wt) plasmids. (B) S. epidermidis RP62a and the isogenic Δcrispr mutant LAM104 were transformed with a mix containing equal amounts of pnes(wt-d) and pnes(mut-d) or pnes(wt-i) and pnes(mut-i) DNA, as indicated at the bottom. Transformation efficiency was calculated as cfu/μg DNA. (C) Primers P86 and P87 were used to amplify a 553 bp PCR product from 10 colonies (lanes 1-10) obtained in each of the transformations shown in (B). DNA was cut with SphI and separated by 1.5 % agarose gel electrophoresis. Purified pnes(wt) and pnes(mut) plasmid DNAs were used as controls. pnes(mut-d) (gels A and B) PCR product digestion generates two fragments of 222 and 331 bp. pnes(mut-i) (gels C and D) PCR product digestion generates two fragments of 328 and 225 bp. 7

9 Supplementary Table S1. Primers used in this study. Primer Sequence P1 ggggacaagtttgtacaaaaaagcaggctcaaaattattattgaaaggtacagg P2 AaaGGTACCgatactttaacaaatgccatcac P3 aaaggtaccattgtagattttgaataaaatacgc P4 ggggaccactttgtacaagaaagctgggttttcttcataacatctaattagcg P6 AaaGGTACCaaatttaatgctattttccttcgc P13 Gtggtgagaatgatggatgag P14 Tctcatgtttccgatccacgc P15 ggggacaagtttgtacaaaaaagcaggcttgaagatagattaaataaaattgagg P16 Cttatattggtgattaatgtattttggcatg P17 Catgccaaaatacattaatcaccaatataag P18 ggggaccactttgtacaagaaagctgggtgttatttaagtggctgggggc P19 aaaggtacctttaaagtatatatcagattgtttcg P27 Aaagtaatttatctagctggcc P28 Tcaaacaattaggccacttcc P37 aaaggtacctattttttgacagcaaaaatgatgc P51 cattactgtataaagatacaattatttatatacttcggcatacgttctc P52 gagaacgtatgccgaagtatataaataattgtatctttatacagtaatg P53 gattcttacctttgtactgatgatttcaattatgttacgaataggttcg P54 cgaacctattcgtaacataattgaaatcatcagtacaaaggtaagaatc P67 Gtgacctccttgccattgtc P68 Tttgtactgatgatttatatac P69 Acgtatgccgaagtatataaatc P70 AaaaAAGCTTcaagaatccaatgaagtagggg P71 aaaaaagcttctaaattagaacatgatactaacg P72 AaaaAAGCTTatcaacaaaatggccaacgatc P76 Gagaaagtgcaactattccgataggaagtaggg P77 Ccctacttcctatcggaatagttgcactttctc P86 Catatagttttatgcctaaaaacc P87 Atatatttatttggctcatatttgc P156 Ctaacagcacaagagcggaaag P157 Acatcaaatcttacaaatgtagtc (*) Capital letters indicate the sequence of the restriction site inserted. Bold capital letters indicate att sites for Gateway cloning (Invitrogen) into pkor1. 8

10 References for Supporting Online Material S1. S. R. Gill et al., J. Bacteriol. 187, 2426 (Apr, 2005). S2. Y. Q. Zhang et al., Mol. Microbiol. 49, 1577 (Sep, 2003). S3. B. N. Kreiswirth et al., Nature 305, 709 (Oct 20-26, 1983). S4. M. Kaltwasser, T. Wiegert, W. Schumann, Appl. Environ. Microbiol. 68, 2624 (May, 2002). S5. O. Schneewind, P. Model, V. A. Fischetti, Cell 70, 267 (Jul 24, 1992). S6. T. Bae, O. Schneewind, Plasmid 55, 58 (Jan, 2006). S7. R. Barrangou et al., Science 315, 1709 (Mar 23, 2007). S8. S. J. Brouns et al., Science 321, 960 (Aug 15, 2008). S9. H. Deveau et al., J. Bacteriol. 190, 1390 (Feb, 2008). S10. P. Horvath et al., J. Bacteriol. 190, 1401 (Feb, 2008). S11. M. Landthaler, D. A. Shub, Proc. Natl. Acad. Sci. USA 96, 7005 (Jun 8, 1999). S12. S. Horinouchi, B. Weisblum, J. Bacteriol. 150, 815 (May, 1982). S13. W. M. Reeves, S. Hahn, Mol. Cell. Biol. 25, 9092 (Oct, 2005). S14. Current protocols in molecular biology. F. M. Ausubel et al., Eds. (John Wiley and Sons, Inc., New York, ed. 1st, 1987), pp. S15. T. M. Morton, J. L. Johnston, J. Patterson, G. L. Archer, Antimicrob. Agents. Chemother. 39, 1272 (Jun, 1995). S16. S. Schenk, R. A. Laddaga, FEMS Microbiol. Lett. 73, 133 (Jul 1, 1992). S17. B. A. Diep et al., Lancet 367, 731 (Mar 4, 2006). S18. M. W. Climo, V. K. Sharma, G. L. Archer, J. Bacteriol. 178, 4975 (Aug, 1996). S19. T. Berg et al., J. Bacteriol. 180, 4350 (Sep, 1998). S20. L. M. Weigel et al., Science 302, 1569 (Nov 28, 2003). 9